基于遗传退火算法的质检扰动应对方法  

Quality inspection disturbance response method based on genetic annealing algorithm

作　　者：葛艳 王爱民[1] 叶介然 GE Yan;WANG Aimin;YE Jieran(School of Mechanical Engineering, Beijing Institute of Technology, Beijing 100081, China)

机构地区：[1]北京理工大学机械与车辆学院,北京100081

出　　处：《计算机集成制造系统》2021年第11期3159-3171,共13页Computer Integrated Manufacturing Systems

基　　金：国防基础科研计划资助项目(JCKY2017602C015,JCKY2018208A001,JCKY2018203B009,JCKY2018204b016)。

摘　　要：针对工序质检结果使原作业计划不能有效指导车间生产的现状,研究考虑工序质检的柔性作业车间动态调度问题。建立以最小化工件的最大完工时间和最小化排产方案变更差异为目标的混合整数规划模型,并提出一种基于局面评价的遗传退火算法。该算法将遗传算法的种群和变异概念引入模拟退火算法,利用模拟退火算法多次获得局部最优解以及大规模变异跳出局部最优的机制,获得最终全局近优解。在解码规则中直接考虑调度目标,提出基于局面评价的解码机制,避免产生劣质解,从而减小解空间。采用所提算法对文献中的案例进行扩充和求解,并与3种算法对比,验证了所提算法在解决该类问题上的有效性和优越性。For the current situation that the original scheduling scheme lost guidance to workshop production caused by the result of operation quality inspection,a dynamic flexible job-shop scheduling problem with quality inspection operations was studied.Aiming at minimizing the makespan(Cmax)and the differences between the original and updated schemes,a mixed integer programming model was established.To solve the model,a genetic annealing algorithm was proposed.In this algorithm,the concepts of mutation and population of traditional genetic algorithm were introduced into the traditional simulated annealing algorithm.By using the simulated annealing algorithm,the local optimal solution was obtained repeatedly,and the mechanism of large-scale mutation to jump out of the local optimal solution was also acquired,thus the final global near optimal solution was obtained.In addition,considering the scheduling objectives directly in the decoding rules,a decoding mechanism based on situation evaluation was proposed to avoid the generation of poor solutions and reduce the solution space.A software system for algorithmic comparisons was developed to verified the validity of the proposed algorithm.

关 键 词：柔性作业车间动态调度问题 质检 遗传退火算法 局面评价 

分 类 号：F406[经济管理—产业经济]